Get Insights to Drive the Energy Transition with Next-Gen APM

As the energy transition charges forward, organizations have to do more than avoid unplanned downtime and reduce O&M costs. Advanced Asset Performance Management (APM) can help manage the mounting demands and abundance of data in industry.

With more data than ever at our disposal, it is becoming increasingly difficult to cut through the noise. Now, organizations are tasked with making legacy assets more efficient, on-boarding renewable assets, meeting decarbonization goals and are met with more options than ever before to achieve those goals. With Asset Performance Management (APM), organizations can leverage data to perform advanced workloads, deploy prescriptive analytics, and enhance efficiency with a composable and interoperable SaaS solution.

APM, when used correctly, is a powerful tool that can dramatically increase the reliability and availability of assets, while also helping to reduce the overall operational and asset risk profile of an organization. Today, APM is designed to provide users the ability to assess and assign asset criticality, develop maintenance strategies, perform condition-based maintenance, deploy AI/ML models for near real-time monitoring, support detailed reports and asset integrity, and even utilize Digital Twins to help ensure optimal thermal performance. While currently capable of impressive outcomes, some APM solutions fall flat in enabling innovation to address emerging requirements.

In heavy industry, assets are designed and built to have long lifespans and are optimized to be as cost effective as possible. However, as time passes, these assets could cause potential issues for operators that include technological obsolescence, increased maintenance costs, limited integration and data access ability, and a slew of other potential concerns. Therefore, asset intensive organizations now must optimize these assets to maintain expected performance, increase flexibility and reduce emissions.

Due to the need to optimize asset performance, increased flexibility and reduce emissions organizations are finding it difficult to prioritize against these new targets. Leveraging legacy APM strategies or point solutions no longer allows the innovation required to be futureproof.

To support the continued digitization required to support the energy transition, organizations must continue to consider cost, availability, and emissions reduction for assets. Looking into 2024 and beyond, there are a few critical areas that the next-generation APM must address to enable the success of organizations.

For the next generation of APM it will be critical to:

1. Help manage data strategy, aggregation, and contextualization across systems and into a central location.

Trustworthy data from all assets’ data sources is a prerequisite for the more advanced APM capabilities of the future. Improving data volume and quality will allow for advanced artificial intelligence (AI) and machine learning (ML) techniques, as well as help companies standardize data across their enterprise to maximize value from new insights. For example, the next-gen systems will not only use maintenance data (IT system) and asset health data (OT system), but also data from engineering models (ET system) that can leverage detailed asset data and visualizations, such as 3D engineering models, so that outcomes and insights can be realized faster, with higher level of confidence. 

In order to also address energy transition requirements, APM will also need to leverage data from other generating assets. Next-generation APM systems will need to integrate with or include asset and fleet management capabilities, distributed energy resources (DERs), and renewable sources. Data such as plant/facility capacity, time to load or bring on-line, cost, contracted sources, demand/ expected demand, and load forecasts, emissions, and transmission losses, when brought together, give you a truly cohesive overview of how assets and the fleet are running. Advanced APM must support transmission and distribution (T&D) and renewable and other emerging asset types to truly generate a data ecosystem focused on key objectives.

2. Further leverage AI/ML and Digital Twin technology across new use cases to amplify human and asset performance.

AI/ML is often discussed, but rarely boiled down to specific use cases. Software providers view AI/ML as a way to generate autonomous workflows, eliminate human intervention, and increase overall operational efficiency. For APM, however, AI/ML and Digital Twin technology needs to be deployed with intentionality to usher in the next generation of performance.

With more data than ever to manage, AI/ML in next-gen APM allows predictive models and Digital Twins to be even more powerful in analyzing the state of assets and facilities. This technology, although it gets smarter over time, must be used to amplify the expertise of your employees on the ground, meaning that AI/ML must also be able to consider human inputs more efficiently and allow intervention.

To optimize for the energy transition, AI/ML within APM must move beyond traditional asset failure modes managing to “high availability” alone. Since operators can’t just “turn off” older assets, they will need more advanced advice on how to run older assets in line with new carbon neutral goals. This requirement opens the door for organizations to generate prescriptive recommendations with models that assist in providing end users automated insights from previous outcomes.

Many Digital Twins in use today in APM systems are excellent in analyzing the operation of assets from the standpoint of traditional failure modes and providing recommendations on how to keep the asset running at peak efficiency. But what if the goal must change from “peak” efficiency to “carbon management” efficiency?  Digital Twins will require new inputs that account for carbon intensity. These inputs, coupled with the other comprehensive data sources, can provide recommendations that balance cost, risk, availability, and carbon intensity.

3. Limit rogue development, while still supporting innovation.

As organizations grapple with the increase of data and integration of AI/ML supported by APM, a new trend emerges: rogue development. Rogue development in open software ecosystems is uncontrollable if no development standards are put in place and monitored. For example, a large Oil & Gas organization has APM deployed across up, down, and midstream. In up and downstream, users have shared standards and focus on composability. However, in midstream, users are generating custom models on top of APM to accomplish similar tasks. This practice leads to future technical debt, increased cybersecurity concerns, and dilutes potential outcomes.

Due to this, the next generation of APM must allow a high level of composability including advanced visualizations, analytic creation, advanced policies and integrations. To support organizations in their pursuit of critical performance measures, APM must give users access to tools that can be configured to their specific use case and provide parameters for future development. By moving towards configurability, organizations can be more agile, scale APM quickly, reduce IT costs associated with development times and infrastructure costs, increase interoperability, and make maintenance and personalization much easier.

By considering the new data sources APM will have to ingest and focusing on how to align technology like AI/ML, APM will be required to evolve to meet the business demands of tomorrow.

4. Allow for various deployment types with a focus on interoperability.

2023 saw a massive acceleration of cloud-deployed APM, along with uncovering pain points in the space as it pertains to on-premises modernization or hybrid hosting options. As organizations require more data sources from across the enterprise to support business objectives, interoperability will become more critical for APM. Interoperability provides immense technological benefits including the ability to exchange data in a standardized format, integrate across platforms, enable plug-and-play for future applications, and help to integrate legacy systems and data into new instances.

But, with the increase in interoperability, organizations will likely be working across on-premises and cloud applications of varying infrastructures and customizations. Therefore, APM in the future will need to be able to operate in the cloud, on-premises, or a combination of both (hybrid). Hybrid cloud is an emerging strategy that was spawned primarily from the needs of financial institutions to modernize their software stack. Now, the need to modernize is extending into asset-intensive industries as well.

To provide the next generation of APM, data must be able to be easily moved and action across deployment types and help remove the siloes created from years of software additions.

The advancement of the energy transition demands new approaches to APM. We must go beyond optimizing assets, plants and fleets for cost and longevity, and look at what operations and maintenance choices we can make to reduce carbon emissions while balancing risk.

Next-gen APM empowers companies to manage systems optimized for reducing emissions while hitting other business goals around cost-effectiveness, availability, and risk. By bringing in more robust qualified data, interpreting it using Digital Twins, and presenting it in intuitive frameworks that requires less specialized knowledge, you can make decisions needed to accelerate energy transition.

To do so, APM needs to ensure that new data sources can be easily on-boarded and integrated. This means a focus on data loaders, application programming interfaces (APIs), and other strategies that can allow organizations to easily bring in new assets, sensors and other data sources as seen fit.

Next, APM must be able to provide intuitive AI/ML technology that amplifies human performance. In high-risk environments and asset-intensive industries, workers require real-time data and context to support the organization’s goals. From there, organizations that are operating at scale across data sources need an APM that is composable based on requirements. Composability gives users the setup they need, but also avoids rogue development which could increase development costs. This approach still allows innovation but helps ensure consistency.

Finally, to bring all the pieces together, APM needs to be flexible in how it can be deployed. This means APM must work across cloud, on-premises, and hybrid data sources to support critical workloads. With this flexibility, organizations have a powerful APM that can truly act as the single pane of glass for all asset-related data.

As we look forward beyond “next-gen APM,” the future will require knowing what generating capacity is available across a fleet of assets, and then accounting for emissions from all sources be they traditional, microgrids, energy storage, vehicle to grid (V2G), solar, wind or biomass. It will require tighter integration to supply chain processes and systems to use asset management outcomes to augment decisions around the criticality of assets that drives inventory management and procurement costs.